About TheJokeDB

TheJokeDB is a collection of jokes you can actually tell, and every one of them has been read by a real person before it was allowed in. That is the whole idea. The internet has more jokes than anyone could read in a lifetime, and most of them are either not jokes, not funny, or not something you could say out loud at a table with your in-laws at it. We read them so you do not have to.

Jokes arrive from people who submit them and from public archives. We use AI to help classify and categorise them: a type, a topic, a rating from clean to explicit. A person reads every joke, fixes what needs fixing, and decides whether it goes live. Duplicates are caught by the text, not the title, so the same joke told six ways becomes one joke.

Family mode is on from the start, which means the site shows only clean, all-ages jokes until you flip the switch. Mild jokes (innuendo, light language) are one flip away; the crude and explicit shelves are behind a filter on the browse page, and they never show up in search results or listings unless you ask. We think that is the right default for a site people open in front of their kids.

The site is built to be read, not scraped: the browse page is a map of every topic, mood and type, the topic pages show a daily selection, and the shelves deal one joke at a time. Developers can use the JSON API, which is free for single jokes and search and keyed for bulk listings. If you spot a joke that should not be here, a credit that is wrong, or a typo, the contact page reaches a human.